The proposed research is based on a perceptual phenomenon described by McCollough in 1965 (now known as the McCollough effect). The McCollough effect is readily demonstrated. During an induction phase, an observer inspects two chromatic patterns that alternate every few seconds for several minutes. For example, one pattern could consist of black horizontal bars on a green background and the other pattern could consist of black vertical bars on a magenta background. Subsequently, complementary color aftereffects contingent on the orientation of the bars are noted. In the above example, the achromatic background of vertical bars appears greenish and the achromatic background of horizontal bars appears pinkish. A striking characteristic of the McCollough effect is its longevity. The illusory colors are seen days, and perhaps weeks, after a short induction period. The proposed research has two objectives: (1) to evaluate the extent to which contingent color aftereffects (which result from pairing of pattern and color) can be understood as an instance of another well-studied phenomenon of pairing, Pavlovian conditioning, and (2) to study a new contingent aftereffect: illusory color seen on achromatic words following exposure to these words in color. The research concerning the contribution of Pavlovian conditioning to the McCollough effect will aid in understanding the contribution of learning to perception. Specifically, these experiments are designed to evaluate the extent to which the longevity of the aftereffect (and other characteristics of the phenomenon) can be explained by learning principles. The research concerning the word-contingent color aftereffect may aid in understanding how people code and process the written word. In addition, it is likely that the word-aftereffect research is relevant to reports of troublesome contingent color aftereffects suffered by computer operators. These aftereffects have been implicated in ophthalmologic, psychiatric, and other work-related problems of computer operators. In all experiments, a computer system will be used for induction and measurement of contingent color aftereffects. Modified versions of standard psychophysical procedures (method of adjustment and method of constant stimuli) will be used to objectively measure aftereffect magnitude.